JPS63215166A - Area processing system - Google Patents

Abstract

PURPOSE:To save the storage capacity and the transmission time of an area data and to facilitate the editing of a complicated irregular area, by dividing the irregular area where plural rectangles are combined into rectangular areas. CONSTITUTION:When first point P1 (PX1, PY1) and second point P2 (PX2, PY2) are inputted, since it is ¦PX2-PX1¦>¦PY2-PY1¦, the input point P2 is corrected to T2=(TX2, TY2)=(PX2, PY1). When third point P3 (PX3, PY3) is inputted, it goes to ¦PX3-TX2¦<¦PY3-TY2¦ from a corrected second point T2 and the third point P3, and the P3 is corrected to T3 (TX3, TY3)=(PX2, PY3). Fourth point P4 (PX4, PY4) is compared with a corrected third point T3, and is corrected to T4 (TX4, TY4)=(PX4, PY3). Fifth point P5 (PX5, PY5) is corrected to T5 (TX5, TY5)=(PX4, PY5) and sixth point P6 (PX6, PY6) to T6 (TX6, TY6)=(PX6, PY5), and the first point P1 (X1, Y1) is corrected to T1 (TX1, TY1)=(PX6, PY1).

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for processing a region input by a coordinate input device in an image processing apparatus.

[Prior art]

Conventionally, one of the methods for specifying an area is to specify a rectangular area using two end points on the diagonal line, and furthermore, the area to be specified is a combination of rectangles (hereinafter referred to as an uneven area), and one corner point of the area is specified sequentially. There is a way to enter it.

In such an input method, a method is used in which the change points in the horizontal and vertical directions of an area are used as boundaries, and the content of image processing is switched.

However, it has been difficult to perform image processing on a region that is a complex combination of multiple rectangles.

[Problems to be Solved by the Invention] Therefore, as a method of performing image processing on a region with a complex shape, a method of having a bitmap memory for editing work can be considered. As shown in FIG. 15, the inside of the specified area is written into the memory corresponding to the image reading area, and the output is controlled while referring to this bitmap memory. This method has the advantage that it is possible to easily process multiple areas that undergo different image processing. However, on the other hand, there is a drawback that the memory capacity required for the bitmap memory is very large. As an example, an image area of A4 (296 x 210) has a resolution of 0, 5 mm x 0, and 1
If you try to process 1 image, you will need about 250,000 bits.

Furthermore, it is difficult to perform this processing in real time.

〔the purpose〕

SUMMARY OF THE INVENTION In view of the above circumstances, it is an object of the present invention to provide an area processing method that allows image processing to be easily performed even on areas with complex shapes.

〔Example〕

An embodiment of the present invention will be described in detail below with reference to the drawings.

1-2 are external views of a copying machine to which the present invention can be applied. The document reading unit 101 includes a document table glass (not shown), a 2:1 optical system consisting of an illumination lamp and a mirror, a lens, and a CCD.
The optical system performs sub-scanning while photoelectrically converting the original image and outputs it as an electrical signal. The printing unit 102 is a so-called laser printer, which receives an electrical signal from the reading unit 101, scans the photoreceptor with a modulated laser beam, and prints using an electrophotographic method. The reading unit 101 includes an operation unit 103, a coordinate input device 104 that also serves as a document holder, and a coordinate input pen 105. FIG. 1-1 shows a view of the coordinate input device 104 from above. 106 is an effective coordinate input surface, 107 is a reference point for placing the original 114 on the input surface, 108 is an area specification key for inputting an area specification, and 109 is a clear key for canceling a series of input coordinates. , 110 is an OK key to finish inputting the area designation,
112 is a masking mode selection key that erases only the specified area; 113 is an image separation mode selection key that outputs the inside of the specified area with different processing from the outside.

Examples of these operations will be explained.

As shown in the figure, the operator places the original 114 against the reference point 107, enters the area designation mode using the key 108, and uses the input pen 105 to move clockwise or counterclockwise around each corner indicated by a black circle in the area surrounded by a dotted line. After inputting data sequentially by rotating the keys, press the OK main key 10 to complete the input. After that, 111. 11
2. 113 keys.

FIG. 3 shows a schematic block diagram of the copying apparatus of this embodiment.

302 is an image reading unit. The image reading unit 302 is a CCD
, a signal amplification circuit, an A/D conversion circuit, and a shading correction circuit, and the corrected signal is input to the image processing unit 303 .

The image processing unit 303 once stores the image signal in a shift memory, performs image processing such as scaling, movement, trimming/masking density conversion, etc., and outputs the result to the external unit 312 as a VIDEO signal. In this embodiment, the external unit 312 is assumed to be a so-called laser printer, but it may also be a controller and memory device for an electronic file, facsimile, etc.

Image signal VID is connected to the external unit via cable 313.
In addition to EO, a serial signal line 5RAL consisting of a plurality of signal images is used to send a serial signal to control the entire system. When the external unit 312 is a laser printer, it receives a horizontal synchronizing signal BD from the printer, generates a clock for internal operation by a clock generator 304, and supplies it to the reading section 302 and processing section 303 described above.

The CPU 301 includes a ROM 305 that stores control programs and control data, and a RAM 3 that stores processing data, etc.
06, has a timer circuit 307, and the above-mentioned reading section 302
) In addition to controlling communication with the processing unit 303 and external unit 312, it also controls a fluorescent lamp driver circuit 309 for illuminating the document, a motor driver circuit 308 for driving the optical system, the operation unit 311, and the coordinate input device 310. .

FIG. 4 shows a block diagram of the coordinate input device 310.

Timing circuit 402 generates a clock using a built-in oscillator and supplies it to address counter 403 . The address counter 403 performs a counting operation while the switch built in the pen 412 is pressed down. During the counting operation, the X decoder 409 operates first and dynamically scans the pulse voltage to the X-direction electrodes of the input surface 411 twice, and then the Y-decoder 410 operates to sequentially apply the same pulse voltage to the Y-direction electrodes of the input surface 411. Performs two cycles of dynamic scanning.When the pen switch is turned on on the input surface 411, when the scan in the X and Y directions reaches the pen position, a voltage is induced in the pen by electrostatic capacity coupling, and the pen input signal is is obtained and amplified by the amplifier 404.The count value is set in the X register 406 by the pen input signal in the first round of the X direction scan, and the contents of the X register 406 and the count are set by the pen input signal in the second scan. The values are compared by a comparator 405, and if they match, a matching signal is output to the CPU 401.Similarly, 1
During the second scan, the Y register 408 is set, and during the second scan, the comparator 407 compares and if they match, a match signal is output to the CPU 401, and when the CPU 401 obtains two match signals, x, The coordinate values can be taken in from each Y register and notified to the main CPU 301 by, for example, serial communication. Note that if the X register 406 and the counter value do not match, it is determined that the input is invalid. The same applies to the Y direction.

FIG. 5 shows a circuit diagram related to the inside of the image processing unit 303, particularly the shift memory. Although the shift memories are provided for two lines, their control is common, so FIG. 5 shows only one shift memory.
The control configuration is shown below. A write address counter 904 is an address counter for writing data into the shift memory 907, and a read address counter 905 is an address counter for reading data from the shift memory 907. The address selector 906 receives a command from the CPU 301 via the I10 port 901, selects either the address signal of the write address counter 904 or the address signal of the read address counter 905, and addresses the shift memory 907. .

I10 registers 902 and 903 are registers for the CPU 301 to give preset values to the write address counter 904 and read address counter 905, respectively.

Both the write address counter 904 and the read address counter 905 are down counters, and the WST signal and R5T signal that command the start of counting operation are input to each of them, and the write clock WC to the shift memory 907 is input to each of them.
LK and a read clock RCLK from the shift memory are input.

915 and 916 are exclusive OR gates for determining the image area, and the signal OF is a signal that controls them. The area within the time frame is the output image, and the area outside the frame is masked.

910 is output from the shift memory 907 and sent to the density processing section 9.
08 is an AND gate that controls the output of the image data that has become a binary signal, 917 is an AND gate that determines whether to output the aforementioned mask portion as white or black, and BB is a signal that controls it. When it is 1, white is output when it is 0.

911 is an OR gate that outputs the image output from gates 910 and 917 as VIDEO; 909 is an exclusive OR gate that controls black and white inversion of image data; IN is a signal that controls it; when it is 1, the image is as original;
Invert when . Each signal is output by the CPU 301 according to the mode specified by the operator.

The ST counter 912 and the EN counter 913 are a start bit counter and an end bit counter for outputting an image only to a predetermined area, respectively.
The CPU 301 presets count data for the gate via lo.

The flip-flop 914 is set when the ST counter 912 counts up, and is reset when the EN counter 913 counts up.

For example, in the case of the OF multiplied signal 1, when the ST counter 912 counts up and the F/F 914 Q becomes 1, the gate 915
There is no output from the gate 910 and it is masked until the output of the gate becomes O and the EN counter 913 increments by a current.

Instead, the output of the gate 916 is 1 during that time, so when the BB multiplied signal is 1, the gate 917 is 1, and the gate 911 outputs 1, resulting in a black mask. Conversely, 0F=1. BB=0(7
) is masked white. Also, if 0F=0, gate 91
5. Since each output of 916 becomes 1.0 during that time, BB
When = 1, the outside of the trimming area is black, 0F = 0. BB=O
When , the area outside the trimming area becomes white.

Now, this embodiment is effective when performing editing processing such as trimming and masking on a rectangle or a region that is a combination of rectangles using the above-described means.

A rectangle or a combination of rectangles is an area surrounded only by line segments parallel or perpendicular to the main and sub-scanning axes, and it is quite practical to limit the editing target to such areas. Moreover, the present invention is useful in that editing functions can be provided at low cost without using a large capacity memory.

Based on these limitations, the operator inputs coordinates discretely to surround the desired area, and makes sure that the line segments connecting these coordinates in the input order are as horizontal or perpendicular to the main and sub-scanning axes as possible. Please input as follows.

However, in reality, it is impossible to input two points connected horizontally and vertically using freehand pen input, considering the practical resolution of the coordinate input device. Correct the coordinates.

FIG. 2 shows a schematic example of the area designation method according to this embodiment.

■First, press the clear key to clear all the coordinates that have already been set.

■Input the first point P+ (PXI, PYI).

■Input the second point P2 (PX21 PY2). At this time, comparing 1PX2 PXI l and 1PY2-PYI l, 1 px2-px, 1 > 1py2-py
, +, so the input point P2 is < T2 (TX21
Correct to TY2 )=(P X21py, ).

■Input the third point P3 (PX3.PY3). At this time,
1PX3-TX21 from corrected second point T2 and third point P3
Therefore, T3 (TX31 TY3) = (PX21
Correct as shown in PY3).

■The fourth point P4 (PX4.PY4) is also compared with the corrected third point T3 and T4 (TX4.TY4) = (PX4.PY3)
and correct it. Similarly, ■The fifth point P5 (PX5 + PYs) is T 5 (T
X s , TY 5) = (PX4.PY5) and ■ 6th point Pa (PXa , PYe ) is T6 (TX6
．． TY6) = (px6.py5).

■When inputting each corner of the desired area is completed, press the OK key.

■When the OK key is input, the first point P, (PX,, PYI
) and the corrected final point Ta” (TXa + TYa)
From = (PXa + PY5), the first
Point P.

(X, , Y,) as T, (TX, , TY+) −
(PXa.

PY, ).

By correcting and controlling the input coordinates in the above-described procedure, the area desired by the operator can be constructed of vertical line segments.

FIG. 6 shows a detailed flowchart of the above-mentioned correction process. First, a counter i on the RAM indicating the number of input coordinate points is cleared to zero (601). If there is a coordinate input (602), set the X component to the area PXi, TXi on the RAM and the Y component to the same (PYi, TYi) (603
).

If i=0, that is, the first fish (604), the counter i is incremented to wait for the second fish input (615).

If i≧1 (604), the corrected coordinates of the previous point (T
i-+.

TYi-+) and the current input coordinates (PXi, PYi).

The absolute values α and β of the difference for each Y component are determined (605).

The operator compares the magnitudes of α and β based on the premise that the line segment connecting each point is input so as to be as horizontal and perpendicular to the axis as possible (606). If α<β, the operator is Y
It is determined that an attempt was made to input coordinates parallel to the axis, and the Y coordinate PYi of the point of interest Pi is adopted as TYi,
The X coordinate PXi is canceled and the previous X coordinate T
i-+ is adopted as TXi (608). On the contrary, α〉β
If so, it is determined that the operator has input the coordinates parallel to the X axis, and the X coordinate PXi of the point of interest Pi is set as T.
Xi', cancel the Y coordinate PYi, and use the previous Y coordinate TYi-+ as TYi (6
07). With the correction of 607°608, the input point Pi becomes Ti
After the line molecule i-+Ti becomes parallel to the X-axis or Y-axis, check whether the line molecule i-+ Ti and the previous line molecule 1-2Ti-+ are orthogonal to each other. investigate.

If i = 1, only one line segment is generated, so the counter l is incremented by 1 to wait for the input of the third point (60
9.610.615).

If i≧2, T X i −2: T X i, that is, whether the two line segments of interest are on the same line segment parallel to the Y axis (
612) or TYi-2=TYi, that is, whether the two line segments of interest are on the same line segment parallel to the X axis (611
), and if No, the counter i is incremented by 1 to wait for the next input (615). If YES, combine these two line segments into one line segment (Ti-2TI-1+
Replace the sum vector 1=r with T I −2T r−, + (613, 61
4). In other words, the previous input Ti-+ is canceled and Ti
Since it treats as Ti-r and waits for T1 human power again, it does not increment the counter i and waits for the next coordinate. After incrementing the counter i to wait for a new input, the line molecule i −2T generated by the latest input point Ti−+
i-1 is all line molecules input so far. T1. T
, T2...Ti-3] is checked (616). The detailed flow will be described later.

If the result of intersection detection is that there is no intersection at all, the next coordinate input is waited (617, 602). If it intersects more than once, Ti
-+ is invalid input and counter i is set to 1 to wait for re-input.
Decrement (617→619). If the points intersect only once, at least one closed area has been generated, so the operator waits for the OK main key force to end the area specification or the clear key force to cancel the point Ti-+. If there is a clear key available, the counter i is decremented by 1 and waits for re-input (618→619).

If there is an OK primary key force, proceed to the end point processing described later (62
4).

When there is no coordinate input in step 602, if there is a clear key input, the counter i is decremented by 1 in order to cancel the latest input point Ti-+ and wait for re-input (62
3). If there is an OK primary key force (624), the process proceeds to the end point process described later (624). When the area is determined as a result of end point processing, the process ends, and if it is not determined, the process waits for coordinate input or clear key input again.

FIG. 7 shows a control flow of intersection detection, which will be explained below. First, a counter k on the RAM indicating the number of intersections is cleared to O (701). When i<4, at most only four points, that is, only three sides have been input, so the detection process ends. i″25
When , i- is written to the counter j on the RAM for cross checking.
4 is set (703). After that, the point Ti-+, T
Values A, B using the coordinates of i-z, Tj, Tj-+.

Find C and D (704, 705, 706, 707).

The following function F is used to calculate the value.

F (Tm, Tn) = -1 + TXm = TXn or TYm = TYn: 0: TXm < TXn and TYm
When <TYn 1: TXm<TXn and TYm>TYn 2: TXm>TXn and TYm>TYn 3: TX
The number of hours F for m>TXn and TYm<TYn is 2 points Tm,
The positional relationship of Tn is classified into five cases and values are assigned, which is illustrated in FIG. In other words, when the vectors Tm and Tn generated from the two points Tm and Tn are parallel to the X or Y axis, the value is "-1", when it is upward to the right, it is "0", and when it is upward to the left, it is "
This is a function that outputs "1" when the direction is downward to the left, "2" when the direction is downward to the left, and "3" when downward to the right.

If even one of the values A, B, C, and D has -1'', update j because the sides T j −+ T j may touch but do not intersect. (708→7
09).

If “-1” is not present, the value is A, as shown in 7jl to 718 below.

B, C, and D have the values “0”, “1”, “2”, and “0” in that order.
3”.

Check whether the values are rotated in ascending or descending order. FIG. 9 shows an example in which the determination in step 711 is YES.

If you rotate the example in Figure 9 clockwise by 90 degrees, it will become 7.
12. 713. 714 is determined as “YES” 4
This is the positional relationship of the points. In addition, the example in FIG. 9 is expressed as side T i −2
When the positional relationship is line-symmetrically folded back to T i-+, it is determined as YES in step 715, and then the rows are rotated counterclockwise by 90 degrees (and "Y" is determined in each step 716, 717, and
ES”.In the above eight cases, the counter k
is incremented by 1, and the check is continued until the side T I -2T i -+ becomes j<O at step 710.

When j is decremented by 2 in step 709, the side Tj −+ T3 when decremented by l is the side Ti −2
This is because it is parallel to Ti -1.

The end point processing flow is shown in FIG. 12 and will be explained below.

First check the counter i (1001), if i<2
If so, since there are only two input points at most, the area is not determined (1012), so the process must wait for point input again (end of process).

If i≧3, the counter i value is decremented by 1 so as to indicate the latest input point number (1002).

Next, start point T. vector T from to the second point T1. T.

Determine the direction code Do of 100 based on FIG.
3), the direction code D i-+ of the vector n from the point T i-+ heading toward the latest point Ti is determined in the same way (1
004). Then, according to the end point processing type table shown in FIG. 13, the two direction codes DO and Di-+ and the point T
Coordinates TXo and TY0 of o and coordinates TXi of point Ti.

Select the end point processing type from the positional relationship of the four coordinates of TYi and perform the processing according to the processing contents listed in Table 1. 1 Figure 14 shows an example of correction based on the processing contents of each type listed in Table 1. show. The side TiT is determined by the starting point (Sp) To and the ending point (Ep) Ti newly determined by performing the processing in Table 1.
If o is generated and is perpendicular to the sides rT and Ti-rTi, there is no need to reprocess it, but if it is not, the process below step tool is performed again using the corrected To, Ti, and i. (1006→1001).

After securing the coordinates of each corner that can constitute a region using the above procedure, apply the above-mentioned intersection detection logic to each side to check whether the line segments formed by sequentially connecting each point from To to Ti intersect with each other. Correspond and execute (1007), if even one point intersects, it is assumed that the area is uncertain and waits for the O point to be input again or for the clear key to be manually canceled (1008)
→1012).

In step 1009, among the points from point To to point Ti whose Y component shows the minimum value, the point Ts whose X component is the smallest is set as the area origin (1009).

Furthermore, the area origin Ts is set as TO again, and the sequence of points is traced clockwise from the new To to T I +T2
＋ ..., renumber Ti (10
10), define the area (1011), and end the area specification.

Step 1009. Processing examples of 1010 are shown in FIGS. 11-1 to 11-3.

By the way, in the case of the method of sequentially inputting one corner point of the uneven area, it is difficult to determine whether the change point is the start change point or the end change point of each image process. 16th
When specifying the uneven area shown in Figure 16-1, the information shown in Figure 16-2 is required.

However, if the area is divided into three rectangles and inputted as shown in FIG. 16-3, the information shown in FIG. 16-2 can be immediately obtained. The method of dividing the rectangle will be explained below.

[Rectangular division]

In order to enable image processing by area and processing of uneven areas, it is necessary to convert input data into a set of data of two diagonal end points of a rectangle. Also, from the point of view of memory saving, it is necessary to perform the division optimally.

FIG. 17-1 shows one method for optimally dividing an uneven area. When humans think in this way, optimal division can be performed using either vertical lines or horizontal lines. In this embodiment, a method of dividing along vertical lines will be explained.

The rectangular division process is performed in the 18th. It is executed based on the algorithms shown in Figures 1 to 18-4. The various quantities required for this are shown below.

PBXOO (input data): Contains the Y value and X value of the input point in the order in which they were input. This is the point where the X value is larger in the line segment where the Y value is the minimum value (shown in Figure 17-2).
is the origin and sorted clockwise.

SCD location: This is a work area for setting the attributes (described later) of each input point when searching for paired points (No. 19-1)
Figure 19-2). Note that a pair of points refers to two points forming diagonal corners of a divided rectangle. In this example, 16X16 (bytes
e) Assign.

AREA POINT CNT...Represents the number of input points. Due to the end point processing described above, the number is always an even number.

Here, the attributes of the points are: - All designated points: All input points - Remaining designated points: Among all designated points, those that constitute a paired point. Note that if it is a remaining designated point, it is included in all designated points.

・New candidate points: All points located on the Y value change points in the horizontal line segments that make up the uneven area.

・New decision points: Among the new candidate points, those that constitute paired points. Note that if it is a new decision point, it is included in the new candidate points.

The above four attributes are defined.

When the Y coordinate of is YPi, when YPi-YPi+1>0, the direction flag (PNT D
RF) is set downward. When YPi-YPi+1<O, the direction flag (PNT DRF) is set upward. The direction is determined with reference to the direction flag.

In addition, in this embodiment, the maximum number of points that can be input is 30 points,
As a result, the number of rectangles that can be divided is 15. Therefore, the maximum number of changing points is set to 15 in both the X direction and the Y direction.

Each process performed by each flow shown in FIGS. 18-1 to 18-4 will be explained.

Set to .

First, the pointer (i) is reset so that the pointer points to the aforementioned origin (2-1). Set the direction flag that determines the direction of the point up or down to 1-L (2
-2) Set all specified points and the direction according to the direction flag to the point where the current pointer is.2-3) The point next to the point where the current pointer is located is the remaining specified point (also all specified points).
and sets the direction according to the direction flag (2-4).

Furthermore, it is determined whether there is a change point in the Y direction between the above two points (2-5), and if there is, a new determination point and direction are set for all points (2-6). Finally, pointer counter (i) is incremented by 2 (2-7).

The above operations are repeated for the number of input points (1-8).

First, reset the counter l to 0 (3-1), and if the directions of adjacent new points within the same -Y value are different among those set as new decision points in the above-mentioned SCD TBL 5ETI, that pixel is demoted from a new decision point to a new candidate point (3-2). Then, the counter l is incremented by one (3-3). This is repeated until the maximum value of the coordinates of the SCD location reaches 15 (3-4). Next, the counter l is cleared to O again (3-5).

After that, when looking at the attributes of adjacent points within the same -Y value, (1) there is a new candidate point in the upward direction and next to all the specified points (including the remaining specified points) in the downward direction.

(2) In the upward direction and next to the new candidate point, there are all designated points (including remaining designated points) in the downward direction.

In these cases, the new candidate point is set again as a new decision point (3-6). Then increment the pointer l by one, and! ! Repeat until =15 (3-8).

(3) PAIRPOINT/SCD TBL (7) Search for paired points based on the contents.

The basic principle of paired point search is that ``points that form a pair are remaining specified points or newly decided points, and a point in the upper direction and a point in the lower direction that is the shortest distance to the upper right of that point form a pair.'' .

In addition, in the algorithms shown in Figures 18-3 and 18-4,
In order to be able to handle specified inputs as shown in Figure 17-3,
Extending the conditions for pair point search°, ■ Two points with different directions form a pair, either remaining designated points or new determined points.

■If the point on the left side of the two diagonal points is the starting point and the right side is the ending point, the ending point is the shortest distance diagonally to the right of the starting point. In the Jiang Tong-X value, the diagonally upper right point has priority over the diagonally lower right point.

These two conditions are defined.

In addition, the Y counter with Sk as the starting point, SI! is the X counter at the start point, Ek is the Y counter at the end point, and EI! Let be the end point counter.

First, counter Sk, S! ! 4-1 and 4-2), and search for a point that meets the conditions for the starting point of the two end points of the rectangular area (4-3). The conditions are: 1) a remaining specified point or a newly determined point 2) an unused point (a point that has not yet been registered as a starting point or an ending point since this pair point search started).

Next, find the end point that corresponds to the found starting point (4-4)
.

The method of searching for the end point is to first find the Y coordinate where the end point exists, and then find the X coordinate.

Search for all designated points (including remaining designated points) or a new determination point with the same X coordinate value as the position of the starting point, to the right of the starting point (larger Y coordinate), and with the shortest distance, and use this as the reference point for the end point search. (5-3.5-12). If found, increase the X value by 1 from the reference point position 5-4), ■ It is an unused point ■ The direction is opposite to the starting point ■ A point that meets all the end point conditions of remaining specified point or new determined point Search (5-5).

If the end point is not found, reduce the X value by one from the reference point and search for the end point in the same way (5-6, 5-7, 5-8).

When a point that satisfies the end point conditions is found, that point is determined as the end point and the X coordinates of the start and end points are swapped (5-
9). If it is still not found, increase the Y value of the reference point by 1 (5-10) and start searching for the reference point again in the same way as before. Since input coordinate correction processing and end point processing are performed, one starting point always corresponds to one ending point. Once the starting point and ending point are found, flag these two points as specified, and (5-11) search for the next pair of points.

Through the above processing, the coordinate information of the uneven area is converted into a set of two end points of a rectangle.

Next, a case in which an area as shown in FIG. 20-1 is designated will be specifically explained.

When the rear shown in FIG. 20-1 is specified, end point processing is performed, and data sorting is completed, input items 1 will be as follows.

PBXoo...60, 10, 60, 20, 50, 20
,50,40,70,40,70,60,60,60,
60,70,50,70,50゜50.10,50,1
0, 20, 20, 20, 20, 40, 40, 40, 40
,30,30,30,30.10AREA POIN
T CNT...18 Based on this, SCD TBL
Set.

SCD TBL of the area of Figure 20-1 in Figure 20-2
This figure shows how the coordinates have been converted.

Figure 19-1 shows the state of SCD TBL at the end of SCD TBL 5ETI shown in Figure 18-1, and SCD TBL 5ET shown in Figure 18-2.
The situation of SCD TBL at the end of 2 is shown in the 19th
- Shown in Figure 2.

Pair points are searched based on the SCD TBL in the state shown in FIG. 19-2.

First (2, O) (SCD T shown in Figure 19-2)
A point that satisfies the starting point condition is found at the coordinates on BL).

Then, if you look one place to the right of the starting point, you will find the reference point for searching for the ending point there (2, 1).

Next, search above the reference point and find a point at (5, l) that has the conditions of the end point corresponding to the current start point.Therefore, consider these two points as a pair and register them. shall be.

When we start searching for the starting point again, we find the second starting point at (0, l), and we continue to find paired points in the same way as before.

By performing the bare point search in this manner, a set of rectangular area 2 end point data as shown below is obtained corresponding to FIG. 20-1, and is divided into seven rectangular areas.

For example, by dividing the specified area into rectangles as described above,
In the case of the area shown in FIG. 20-1, what used to be defined by 18 points is now defined by 14 points, which helps save memory capacity for storing area data and time required for transmission. Furthermore, it is also easier to control, for example, the trimming process of such a complicated area.

In order to edit the divided area as shown in Figure 20-1, for example, as you can see on the main scanning line at Y-30, edit the two sections of X = 10 to 20 and X = 30 to 50. For example, it is necessary to perform trimming processing, but in such a case, the counters 912, 913 and gates 909, . 910.915. This can be achieved by having at least two sets of 916.917. Now two sets of counters 912,
913 to ST counter 1. EN counter l and ST counter 2. Assuming that the EN counter is 2, when trimming the area shown in Figure 20-1, when the optical system reaches y=to, S
Set T counter l to 30, EN counter 1 to 60, and ST counter 2. The same value is set in EN counter 2. Next, when the optical system reaches Y=20, ST counter 2
10. 20. to EN counter 2. EN counter 1
Set 50 to .

Furthermore, when the optical system reaches Y=30, 1 is added to ST counter 1.
0. 70 on the EN counter. If the same value is set in counter 2 and the counters are controlled in the same manner thereafter, desired editing becomes possible.

〔effect〕

As explained above, according to the present invention, by dividing a concavo-convex region that is a combination of multiple rectangles into rectangular regions, it is possible to save storage capacity and transmission time of region data, and it is easy to edit a complex concavo-convex region. becomes.

Furthermore, the rectangular division method of the region in this embodiment enables high-speed calculation processing based on the coordinates of each corner of the uneven region and its directional component.

[Brief explanation of the drawing]

Fig. 1-1 is a diagram showing a coordinate input device, Fig. 1-2 is an external view of a copying machine to which the present invention can be applied, Fig. 2 is a diagram showing an example of coordinate correction, and Fig. 3 is a schematic diagram of the entire device. Block diagram: Figure 4 is a block diagram of the coordinate input device; Figure 5 is a processing block diagram related to editing; Figure 6 is a control flowchart for input coordinate correction; Figure 7 is a control flowchart for intersection detection; An explanatory diagram of the box machine F used for intersection detection, Fig. 9 is an example of an intersection, Fig. 12 is an end point processing control flowchart after coordinate input is completed, Fig. 13 is a diagram showing an end point processing type table, and Fig. 14 is an end point processing Figure 15 is a diagram showing an example of bitmap memory for area editing. Figures 16-1 to 16-3 are diagrams showing the effect of rectangular division.
8-1 to 18-4 are the flowchart 1 of rectangular division processing, FIGS. 19-1 and 19-2 are diagrams showing the schedule table (SCD TBL) used for rectangular division, and FIG.
FIG. 20-1 is a diagram for explaining rectangular division, and FIG. 20-2 is a diagram obtained by converting the area in FIG. 20-1 into schedule table coordinates. 104 is a coordinate input device; 105 is a coordinate input pen; 10
6 is a valid coordinate input surface, 912 is a start counter, 91
3 is an end counter. Figure 2 Figure 7 Man! θ Diagram〈Final Shoiri Tire Table〉 ooooooooooooooo. To f6 (AD,) ρノ I'2 (J2.727 guan/7-ba Z' 帛/7-?7 77-3 Ikuan 18-3 Fig.

Claims (4)

[Claims] (1) In an image processing device having a coordinate input device, the coordinate input device converts a closed area specified to be surrounded by horizontal/vertical line segments in each main/sub scanning direction into one or more rectangular areas. An area processing method characterized by having a dividing means and performing desired processing on the divided rectangular areas. (2) In claim (1), the dividing means performs dividing processing based on the coordinates of each end point of a line segment surrounding the closed region and the directional component of a vector generated by sequentially connecting the coordinates of the end points. A region processing method characterized by: (3) In claim (1), the area processing is characterized in that the dividing means divides the closed area into rectangular areas only by line segments parallel to either main scanning or sub-scanning. method. (4) Claim (1), further comprising means for storing the closed area as a set of coordinate pairs of two diagonal points of the rectangular area, and based on the stored coordinate pair set, a desired An area processing method characterized by processing.